THE DREYFUS AFFAIR – FORENSIC DOSSIER OF THE BORDEREAU
Working Papers of Étienne Charavay, Expert en Écriture: Charavay’s Report, Working Notes, & Traced Specimens of Dreyfus’s Own Hand (Paris, October 1894).
The Dreyfus Affair — the wrongful conviction of Alfred Dreyfus, a Jewish officer on the French General Staff, for treason in 1894, and the twelve year struggle to exonerate him — is among the defining episodes of the modern era. It exposed antisemitism operating at the highest levels of the French state and military, revealed how institutional authority could be marshaled to sustain a fabrication against the evidence, and forced a confrontation with the question of whether republican justice could survive the interests of the army that was supposed to defend it. It produced Émile Zola’s J’Accuse...! (1898) — the open letter to the President of the Republic that established the template of the engaged public intellectual, the writer who intervenes in the affairs of the state on the ground of conscience. It split France into irreconcilable camps, generated the modern political vocabulary of left and right as moral rather than merely parliamentary categories, and accelerated the separation of church and state. Dreyfus was exonerated in 1906. The affair has never ceased to be a reference point — for the history of antisemitism, abuse of state power, of expert authority, of the press, and of the relationship between truth and institutions.
The Three Documentary Foundations of The Affair:
Three categories of primary document stand above all others in the history of the Dreyfus Affair. The first is the bordereau itself — the unsigned memorandum recovered in 1894 from the waste-basket of the German military attaché in Paris, which listed several confidential military documents and offered to supply them to the German embassy. Because these documents circulated within the French General Staff, the author was presumed to be an officer with access to its internal papers. The bordereau was the document on which the accusation of treason rested; its authorship was the central question of the case.
The second is the secret dossier: the package of fabricated and manipulated documents passed illegally to the judges in closed session, without the knowledge of the defense, which secured the actual verdict. Its existence was concealed from the public entirely.
The third is the forensic handwriting analysis — the expert examination of the bordereau against Dreyfus’s authenticated writings, which constituted the entire visible public case against him: the only evidence France and the world actually saw, the mechanism by which the bordereau was attributed to Dreyfus, and the evidentiary framework that made the conviction appear legitimate. Several experts produced reports or testimony during the investigation, and their official papers are preserved within French state archives. Yet the authority that made the handwriting case credible belonged above all to a single figure: Étienne Charavay — archiviste-paléographe, expert en écriture près le Tribunal de la Seine, chevalier de la Légion d’honneur, and the most professionally credentialed handwriting expert in the case on either side.
The bordereau is in permanent institutional custody and will never be offered. The secret dossier, dismantled and dispersed across state archives in the course of the revision proceedings, will never be offered as a unified collection. The working dossier of Étienne Charavay, preserved intact since October 1894, is offered here.
The Charavay Dossier:
The handwriting of the bordereau differed markedly from Dreyfus’s authenticated writings. Rather than weakening the accusation, these discrepancies led the prosecution to construct a theory that would allow the case to proceed. Alphonse Bertillon argued that Dreyfus had deliberately disguised his own hand when composing the document — a process he described as autoforgery. According to this theory, the differences between the bordereau and Dreyfus’s known writing were themselves evidence of guilt, the product of a calculated attempt to conceal authorship. The logic was circular and elastic: if the writing resembled Dreyfus’s, that resemblance was evidence of guilt; if it differed, the differences were evidence of disguise. Bertillon’s theory was capable of implicating almost anyone. But once supported by Charavay’s professional authority, it provided the prosecution with the one thing the case required: the appearance of technical proof.
Dreyfus was convicted in December 1894 and sentenced to deportation for life. In January 1895 he was publicly stripped of his rank in a degradation ceremony in the courtyard of the École Militaire — the insignia torn from his uniform, his sword broken before a crowd of soldiers and spectators — before being transported to penal exile on Devil’s Island. The true author of the bordereau, Ferdinand Walsin-Esterhazy, was eventually identified. In 1899, during the retrial at Rennes, Charavay — seriously ill — traveled to court and publicly, solemnly withdrew the conclusion he had presented in 1894. The military court convicted Dreyfus a second time regardless. Dreyfus accepted a presidential pardon under explicit protest, reserving the right to continue fighting for full vindication. A civilian court of appeals exonerated him in 1906.
The Charavay dossier is the physical record of the forensic process at the center of that sequence. It comprises three components:
I. The formal signed expert report, dated Paris 29 October 1894;
II. The preliminary analytical working notes from which the report was composed;
III. A group of handwriting specimens used during the comparative examination, consisting of direct tracings on pelure paper of words and letterforms from the documents in the comparison corpus — every one of which was a writing of Alfred
Dreyfus.
I. THE FORMAL SIGNED EXPERT REPORT. 9 manuscript leaves in Charavay's hand, signed "Etienne Charavay" and dated Paris, 29 October 1894.
Opens with Charavay's declaration of credentials and a commission rogatory from the Prefect
of Police dated 22 October 1894. Identifies thirty pieces examined under oath:
— Pièce n°1: the bordereau (anonymous letter on pelure paper, in-8, 30 lines)
— Pièces n°2–25: authenticated writings of Alfred Dreyfus, comprising official letters to army commanders, notes for the general staff, notes on pelure paper, notes on graph paper, and notes on fine paper
— Pièces n°26–30: five traced letters on pelure paper extracted from Dreyfus's personal copy-book, bearing the copy-book page references 358, 354, 355, and 371
Proceeds through systematic analysis under the following headings: examination of the bordereau; examination of the comparison writings; differences; resemblances; disguised writing; conclusion. Heavily revised throughout — crossings-out and interlinear corrections on virtually every leaf, with particular density in the conclusion section, where an earlier formulation has been rendered unreadable through sustained overwriting.
II. ANALYTICAL WORKING NOTES. 7 manuscript leaves in Charavay's hand (pictured above), organized under four headings corresponding to the structure of the formal report:
— Différences
— Ressemblances
— Mobilité de l'écriture de comparaison
— Pièces de comparaison
Letterform comparisons are reduced to quasi-algebraic notation, with individual pieces referenced by number throughout.
One leaf carries on its verso the heading "notes de la page 16" together with a bibliographic reference to Arthur Chuquet's La trahison de Dumouriez (1891) and a citation from General Valence's letter to the Minister of War — indicating that Charavay used the reverse of sheets from a concurrent scholarly project as working notepaper during the investigation.
III. SPECIMEN TRACINGS. Twelve slips and sheets on pelure paper, keyed to pieces numbered 1, 3, 5, 8, 9, 10, 11, 12, 18, 27, 29, and 30 from the thirty-piece comparison corpus.
The slips are tracings of words and letterforms from the documents under examination, annotated by piece number in black and red inks. Because they are tracings rather than copies, the letterforms they carry are direct transfers of Dreyfus's pen strokes from the original documents, not Charavay's reproductions.
Among them is pièce n°29 — a slip bearing the words:"Mon cher Monsieur Paure / Paris, le 4 octobre 1893 / Provins / Ci-joint” — traced from one of the authenticated Dreyfus letters physically examined by Charavay in October 1894.”
The Expert: Why Charavay was Indispensible:
Among the individuals who gave opinions on the handwriting of the bordereau, only Étienne Charavay possessed the credentials of a true specialist. Alphonse Bertillon, the architect of the prosecution’s theory, was a criminologist famous for developing a system of body measurements for criminal identification — not a handwriting expert. His autoforgery argument was later evaluated by three of France’s greatest mathematicians, including Henri Poincaré, who concluded that its absurdity was so evident it was hard to believe how long the discussion had gone on. The second prosecution expert, Pierre Teyssonnières, was described by contemporaneous sources as an expert of no great repute; he was a civil engineer. The two experts who correctly exonerated Dreyfus — Alfred Gobert of the Banque de France and Eugène Pelletier of the Paris Court of Appeal — were genuine practitioners, but neither held the first rank of formal paleographic qualification.
Charavay’s title of archiviste-paléographe was the highest paleographic credential in France, awarded by the École des Chartes. His appointment as expert en écriture near the Tribunal de la Seine was the premier forensic document post in the country. And his professional reputation rested on a specific, directly relevant achievement: in 1870 he had published the definitive critical study exposing the forgeries of Denis Vrain-Lucas — a criminal who had sold nearly 30,000 fabricated historical manuscripts to a credulous collector, including fake letters from Julius Caesar, Cleopatra, and Mary Magdalene. Charavay was the person in France best qualified to detect a forged document. His endorsement of the prosecution’s case was therefore worth more than all the other expert opinions combined.
Bertillon’s pseudo-science gave the case its elaborate apparatus. Charavay’s name gave it scholarly cover. Without him, the public handwriting case against Dreyfus rested on a crank and a civil engineer. With him, it had the most credentialed document expert in France — the man who had spent his career dismantling exactly this kind of fraud.
The Initial Finding, The Reversal, and The Manuscript:
Charavay’s first assessment of the bordereau, before meeting Bertillon, was that the writing was spontaneous and natural. Those words are not merely descriptive. They are a forensic determination of the first importance, and they demolished the only theory on which Dreyfus’s guilt could rest. A spontaneous and natural hand is by definition not disguised. Disguised writing is effortful, constrained, artificially controlled. If the bordereau was spontaneous and natural, autoforgery was impossible.
Alfred Gobert of the Banque de France — the first expert consulted, whose report was suppressed and who was sidelined before the trial — had reached the identical conclusion: the writing was natural and rapid, excluding disguised writing. Charavay’s initial assessment placed him in precise agreement with the one expert the prosecution could not use. It was only after Bertillon was placed at the disposal of the new experts, specifically to influence their conclusions, that Charavay reversed himself.
The formal report is the physical record of that reversal. Crossings-out and interlinear revisions appear across virtually every leaf, recording successive attempts to force the analytical observations into alignment with the conclusion the prosecution required. The revision density is heaviest in the concluding section. What the backlighting of the conclusion leaf reveals is that the original formulation — written before the final conclusion was set — has been rendered substantially unreadable through sustained, deliberate overwriting. The surviving traces are insufficient for confident reconstruction; the deleted text awaits specialist paleographic examination. What is not in doubt is that Charavay wrote a conclusion, crossed it out with unusual thoroughness, and replaced it with the formulation that stands: that pièce n°1 is from the same hand as the comparison documents n°2 to 30, qualified by the observation that the contrary hypothesis would require admitting a “sosie en écritures” — a coincidence of handwriting so complete as to be, within the restricted circle of persons who had access to the military intelligence in the bordereau, beyond reasonable probability.
That qualification was the last surviving trace of Charavay’s professional conscience. The deleted prior conclusion remains in the document, physically present, awaiting the examination it has not yet received.
The Recantation:
In 1899, seriously ill, Charavay traveled to Rennes and stood up in open court. As Dreyfus himself records in his memoirs, Charavay publicly and solemnly acknowledged his error before the tribunal. He was not correcting a technical footnote. He was withdrawing the scholarly legitimacy that had made the public case possible — the authority of the most credentialed document expert in France, openly retracted before the court that had relied on it. The military court convicted Dreyfus a second time regardless.
Charavay’s recantation was courageous precisely because it was, at the moment he made it, institutionally futile. The parallel he could not have failed to recognize was exact: a man of unimpeachable professional reputation, manipulated into lending his authority to fraudulent documents — the precise pattern he had built his career detecting. Anatole France, godson of Charavay’s father and a close personal friend who had studied the Vrain-Lucas forgeries alongside him, became one of the most prominent Dreyfusard intellectuals. The expert most qualified to recognize a corrupt prosecution had been personally embedded, for decades, in the intellectual network that would eventually expose it. He said what was true. The army convicted Dreyfus anyway. History vindicated him in 1906.
Pièce No. 29 – The Self-Authenticated Dreyfus Tracing:
The most significant element in the specimen tier — and the element that most sharply distinguishes this archive from any comparable holding — is the slip identified as pièce n°29. It reads: "Mon cher Monsieur Paure, / Paris, le 4 octobre 1893 / Provins / Ci-joint." It is one of the five traced letters on pelure paper extracted from Dreyfus's personal copy-book that Charavay describes as pièces 26–30: a direct tracing of Dreyfus's handwriting, transferring his pen strokes from an authenticated original onto the pelure sheet used in the forensic examination.
The authentication argument is contained within the archive itself and requires no external authority. Charavay's report — in his own hand, in the same dossier — cites pièce n°29 by number on two separate occasions. In the analysis of the comparison corpus he observes "deux P majuscules de facture différente aux mots Paris et Provins" in pièce n°29, and "la majuscule M aux mots Mon et Monsieur dans la même ligne." In the resemblances section he returns to the M-form: "Or dans les pièces n°29 et 30 j'ai trouvé cette forme M, qui me semblait si anormale." The words Paris, Provins, Mon, and Monsieur — all four words Charavay names from this document — are the words on the slip. The report and the tracing are in direct, word- by-word correspondence. Charavay has told us, in his own hand, exactly which document he was examining and exactly which words he was looking at.
The Specimen Tracings and the Bordereau’s Hand:
The remaining specimen slips fall into two categories. Slips keyed to pieces 3, 5, 8, 9, 10, 11, 12, 18, 27, and 30 are tracings on pelure paper of words and letterforms from the comparison corpus — authenticated writings of Alfred Dreyfus — annotated by piece number in black and red inks. Because they are tracings, the letterforms they carry are direct transfers of Dreyfus's pen strokes from the original documents: specimens of his hand extracted for forensic use.
The slip keyed to pièce n°1 is of a different character. Pièce n°1 is the bordereau itself: the incriminating document, not the comparison corpus.
The bordereau itself remained in military custody; what Charavay preserved here is its handwriting as filtered through the eye and hand of the examining expert: the analytical reduction of the incriminating document into extractable letterforms, set alongside corresponding forms from Dreyfus’s authenticated writings. The archive thus preserves both sides of the comparison as they were actually examined — the bordereau as analyzed, and the comparison material against which it was measured — held together within a single working dossier.
The report also confirms that pièce n°3 in the comparison corpus bears the name of General Gaston de Galliffet — a central figure in the French military establishment and, as Minister of War under Waldeck-Rousseau (1899–1900), later involved in the partial rehabilitation of Dreyfus.
Historical Importance of The Archive:
The handwriting analysis of the bordereau constituted the central evidentiary basis of the public case against Dreyfus. Of the three primary documentary foundations of the Affair, two are now in permanent institutional custody. This archive—the forensic dossier of the expert whose authority made that case legible—is the third, and the only one that could ever appear on the market.
It speaks directly to the central problems of Dreyfus scholarship. It preserves the formal report, whose conclusion section contains a heavily revised earlier formulation that has not been subjected to specialist study. It includes the pre-rhetorical working notes, revealing the analytical process before it was shaped into argument. And it retains direct tracings of Dreyfus’s hand from the documents physically examined in October 1894, including a self-authenticated slip whose attribution is established within the same dossier.
No comparable primary source is known to have come to market.
PLEASE CLICK ON THE LINK BELOW THE IMAGES FOR A PDF WITH A MORE DETAILED ANALYSIS.
[DREYFUS AFFAIR] DREYFUS, ALFRED; CHARAVAY, ÉTIENNE. Archive of Étienne Charavay. Paris: 1894.16 pages of manuscript notes, 9 page manuscript report (all approx. 6x8in; 15.5x20cm, written in ink); 8 slips of tracings on thin paper (various sizes).
Price: $125,000 .